AI Paper Atlas

Research map / Reinforcement learning / Imitation learning

Action Inference by Maximising Evidence: Zero-Shot Imitation from Observation with World Models

Xingyuan Zhang, Philip Becker-Ehmck, Patrick van der Smagt, Maximilian Karl

NeurIPS 2023 · Reinforcement learning › Imitation learning · approach: Behavior cloning theory and models · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use AIME world-model action inference instead of BCO(0) inverse dynamics modelimitation from observation · V-D4RL random walkerreward/10: 2.11 -> 12.36Improved
Use AIME instead of BCO(0)imitation · V-D4RL cheetah-mixreward/10: 16.08 -> 40.27Improved
Use AIME instead of BCO(0)imitation · V-D4RL cheetah-medium_replayreward/10: 15.47 -> 31.47Improved
Use AIME instead of BCO(0)imitation · V-D4RL distracting walker-total_mixreward/10: 2.15 -> 12.66Improved
Add visual distractions to embodiment dataAIME · V-D4RL distracting Walkerreward/10: 12.66 -> 3.81Hurt
Remove KL term or reconstruction term from the ELBO objective (ablation)AIME · mix -> run, Walkernormalised returnHurt
Use only the KL term instead of both terms (ablation)AIME · mix -> run, Walkernormalised returnMixed
Remove the state-space model (IIDM: forward model on raw observations) (ablation)AIME · mix -> run, Walkernormalised returnHurt
Train an explicit IDM on the latent state instead of implicit gradient-based IDM (ablation)AIME · mix -> run, Walkernormalised returnHurt
Increase number of demonstrations beyond 100 (ablation)AIME · mix -> run, Walkernormalised returnImproved
Use AIME instead of BCO(0) with few demonstrations (2 trajectories)imitation · mix -> run, Walkernormalised returnImproved
Use full-state MDP observation instead of LPOMDP for AIMEAIME · Walker/CheetahperformanceHurt

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy